1.1.43 alpha2 test result

[MWfire]

Config switch is not fully completed yet, only some of the tables can be swapped for now. Switchable tables include ve, lambda, kpa and rpm arrays, spark and boost.

I tested on primepuls >:(
Can you put in config switch injectors setings? I have high ohm injectors for petrol(no pwm) and low ohm for LPG(pwm).

EGOC on lcd is 2.56*ego

[Sprocket]

Again, an explanation that, to me, means nothing.

wen you say 2.56* EGO, what value is the EGO

Do you mean the Rich limit and Lean limit?

I have the lean limit set high at 20% and the rich limit set at 5%. If you multiply 2.56 with 20 it gives the 51 i see on the LCD?

Still doesnt explain what the rich or lean limits are a percentage of

[[email protected]]

Okay, uploaded the firmware and configs, then got the whole AiM datastream.

So do I set lcd_c0=40 to turn it all off?

[Sambas]

So do I set lcd_c0=40 to turn it all off?

lcd_c0 bit 2 is aim enable so that should disable it yes

[[email protected]]

When it is enabled I can't get MegaTune to work, it just sits and gets pounded by the out coming data.

[gunni]

turn megatune off and on again, it should work then.
as the inital startup of the megatune stops the stream,
but you have to disconnect from the port and reconnect with the MT
to get to the unit.

thatÃ,´s how itÃ,´s been working for me.

[Sambas]

and make sure that megatune is set for 19200 speed, in terminal commands 'A', 'D', 'S', 'Q' and "Man" disables the aim stream

[[email protected]]

I'm just thinking of a way of writing a how-to, and disabling the LCD and AiM features in order to get MegaTune up to do various tests seems like a good thing to suggest.

Rob

[[email protected]]

Okay...
There are issues with MegaTune.  When you start up MT in the usual manner the combination of AiM protocol and the fact its now 19200baud causes
My tables are all screwed, every other value in the tables is wrong - I think this is because the wideband learning values are no longer printed out when an mct is done, and they are still in the original tables file.
My ignition and injector outputs were all wrong.  The ignition entries were all blank, the injector entries seemed to contain random numbers.

When cranking I found the RPM was spot on, but when I'd set the ignition and injection outputs correctly the engine would catch and kick back against the starter - this would suggest that the base timing has shifted. My cranking advance is still correct at 4^oBTDC.  Could this be a setting that is incorrect with the new trigger config settings?

Here is my new msq file
http://www.vems.co.uk/VEMSConfigs/SR20DE/SR20DEAfterSomeCleaning.msq

I suggest the following actions:
* Create a UpgradeDefaults.txt configlet that sets the values correctly - notably AiM disabled, switching all new features off.
* Update the make tables to cope with the original and new mct formats.
* Create a suitable .cfg file that defaults to the correct baud rate.
* Write a how-to upgrade document.

I am happy to do the make tables executable as I wrote it in the first place, the rest needs further discussion.

[MWfire]

Okay...
There are issues with MegaTune.  When you start up MT in the usual manner the combination of AiM protocol and the fact its now 19200baud causes
My tables are all screwed, every other value in the tables is wrong - I think this is because the wideband learning values are no longer printed out when an mct is done, and they are still in the original tables file.
My ignition and injector outputs were all wrong.  The ignition entries were all blank, the injector entries seemed to contain random numbers.

Just twice time upload config(upload config.bat) and then tables are right.

[Sambas]

Rob, did you use the config and tables I made? Because there is nothing right with that msq. For example I did set confswitch=00 but in msq it's 5 and Toothwheelwidth is also wrong (i think original value). This looks like either you uploaded your old config or something else gone wrong.

[[email protected]]

I re-downloaded the config and tables files you did for me and found that the final line contained a load of entries without CR+LF.
Editied, remade, uploaded - car started :D

Running very rich, but even so I managed to get some nice acceleration response from an engine that was in all other respects very flat.

[GintsK]

I was able perform some additional tests on real car. Due very high EGT I checked timing for every cylinder - perfect match! Also I check AFR using Tech Edge unit - perfect match! Also playing with EGO control - did not find any lag comparing to 1.1.27.

Engine - old VOLVO B200ET. 60-2. No camsync. RC 550cc injectors using Peak-hold mode. GT28R. G-reddy w-gate solenoid. 1.5bar boost.

[MWfire]

log
http://www.vems.hu/files/DamirMuha/1.1.43/egodatalog200811101429.xls
at 193.9s is throtle stady and 11%. At 194.4s ego start coreting. i don't know why i have 1-1.5s lag.
Here is my config
http://www.vems.hu/files/DamirMuha/1.1.43/megasquirt200811101507.msq

http://uk.youtube.com/watch?v=tg9Fjd0HwYc
On 1.0.73 lcd was much faster

[GintsK]

Are you find EGO correction laggy here? Lambda in this case is close to target and controller do changes slower. If faster response desired, rise up speed limit and lower engine cycles. I do not knew anything about step size: do not see any significant changes in range 1...30  ??? Ego correction here is interrupted by AE too.

My typical settings when sensor is located ~70cm after valves is 64 for speed limit and 5cycles when tuning and 10-15 when already tuned.

LCD is slow at start up. Did you see any lag after that? I don't.